Is Not Provided A Good Thing?

It’s been a few weeks since Google rocked our worlds and stripped keyword-level data from the referrer, so it’s time to examine the fallout and see what’s changed in the world of search marketing. Surprisingly, not much. Keywords were beacons of insight – fantastic beacons – but keywords were a double-edged sword leading us to […]

Chat with MarTechBot

It’s been a few weeks since Google rocked our worlds and stripped keyword-level data from the referrer, so it’s time to examine the fallout and see what’s changed in the world of search marketing. Surprisingly, not much.

Keywords were beacons of insight – fantastic beacons – but keywords were a double-edged sword leading us to create content that unnecessarily repeated specific terms for an ideal keyword density. Writing for a bot rather than for the user is not optimizing for user experience, so Google did something about it. Google head spam-fighter Matt Cutts said it himself: over-optimizing for specific terms now has a negative impact on your SEO.

I believe removing keywords from natural search analytics (note: select keyword data is still available in Webmaster Tools) is a good thing. As graph and native-language search technology improves, we shouldn’t have to manipulate content to highlight specific keywords. If Google can understand synonyms, themes and topics, then a good well-linked article should rank high for all relevant searches.

Maybe I’m the only one who sees this, but having cut my teeth on the paid side of search marketing I’ve been noticing that Google’s interpretation of organic relevance has been on a collision course with AdWords Quality Score for some time.

Focus On Quality Content

Quality Score is calculated when a query matches a keyword being bid on, also known as an exact match query. Broad and phrase match queries do not generate a unique quality score in the AdWords UI; instead, they piggyback on the exact match query instances. Given that the majority of impressions in an AdWords account come from broad- or phrase-matched queries, the entire nature of relevance in AdWords is determined by the relationships between those potentially relevant queries and known keywords.

The idea is to make connections between queries, ads and landing pages to ensure the user experience is relevant and gets the user where they want to go. Seems consistent with PageRank – a little bit mob rule (click-through rate) and a little bit black box.

Now take Matt Cutts’ commentary in context – over-optimization of a page for a given keywords can be detrimental for relevance. Why? Think about how you search. Personally I string together a handful of key terms in no particular order. Take for example a recent search of mine: [running shoes minimalist]. My intent is informational, not transactional, as I’m looking for information about minimalist running shoes.

If Google were to reward pages peppered with the grammatically incorrect and awkward sounding phrase [running shoes minimalist], I would be unhappy with my results because they would appear spammy and low-quality. Clearly, my intent is for a minimalist or barefoot style running shoe, so Google draws connections between the synonymous terms [barefoot] and [minimalist] and delivers the following results:

running shoes minimalist

That’s a highly relevant set of organic results. Even the paid ads do a decent job of returning natural language results in line with my intent.

Focus On Topics & Themes

So what? By removing keywords from the referrer Google is forcing us to focus on topics and themes. It’s a bittersweet outcome because themes can include a lot of keywords, and being optimized for a theme can produce dramatically more search traffic than one specific keyword (a la broad match vs. exact match in paid search); however, more relevant keywords mean more competitors.

Keywords are great, but in the end, being perfectly optimized for a specific keyword pales in comparison to being perfectly optimized for a theme of keywords. Similar to advertisers with high account-level quality scores, I believe publishers deemed optimized for one or more themes will have an easier time ranking for new themes and topics as the evaluation criteria for relevance continues to become more subjective with less specific data to optimize on.

As humans, the more we know about criteria for judgment, the more we will change behaviors to increase the likelihood of a positive result. The only way Google can incentivize publishers to focus on creating quality content without trying to game the system is to remove insight into criteria for judgment — keywords. While annoying from the analyst’s perspective, the impact should be profound and positive for users as publishers and marketers get back to focusing on producing quality content.


Opinions expressed in this article are those of the guest author and not necessarily MarTech. Staff authors are listed here.


About the author

Benny Blum
Contributor
Benny Blum is the Vice President of Performance Marketing & Analytics at sellpoints, the leading online sales orchestration platform, and is based in Emeryville, CA.

Get the must-read newsletter for marketers.